On 10/09/2013 11:36 AM, Olivier Grisel wrote:
> Peter implemented "penalized SVD" with SGD for "Netflix
> challenge"-style matrix factorization problems:
>
> http://code.google.com/p/pyrsvd/
>
> It should be a pretty good baseline to compare performance against.
>
> As for missing data, I would just use scipy.sparse matrices and treat
> non-materialized zeros as missing data for the sake of memory
> efficiency and API simplicity.
>
I remember there was a thread on how to encode missing values, I think 
for the imputation PR.
There were three possible scenarios of how to use the sparse matrix 
structure.
Does anyone have a link?
I thought the recommendation system API was also discussed there.

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to